Vegetation Biodiversity in Coastal Oregon Forests€¦ · Vegetation Biodiversity in Coastal Oregon • In semi-natural forested landscapes, all ownerships contribute to biodiversity.

Vegetation Biodiversityin Coastal Oregon Forests

Janet L. Ohmann and Thomas A. Spies, USDA Forest ServiceMatthew J. Gregory and K. Norm Johnson, OSU

• A new kind of vegetation map

• Uses in CLAMS

• Current vegetation biodiversity

Funding by PNW: CLAMS, Northwest Forest Plan, Wood Compatibility Initiative, Forest Inventory and Analysis

A Novel Way to Map Vegetation

Data from plots(FIA, CVS, BLM, OG)

ClimateGeologyTopographyOwnership

Remotesensing

Vegetationmaps (1996)

Spatial data in GIS

Plot locations

Statisticalmodel

IDNO TREE # SPECIES DBHCM HTM CC BHAGE TPHPLT

41034020 101 TSHE 39.116 24.384 4 83 2.617

41034020 116 CHLA 109.728 32.309 3 136 2.617

41034020 123 TSHE 55.880 39.319 3 103 2.617

41034020 129 PSME 200.152 58.826 3 913 1.000

41034020 133 PSME 66.802 40.843 3 99 2.617

41034020 316 TSHE 57.404 40.234 3 80 2.617

41034020 319 CHLA 105.664 45.110 3 244 2.617

41034020 320 CHLA 80.518 42.062 4 349 2.617

A ‘tree list’ for each pixel

CLAMS vegetation map ...somewhere SW of Eugene, 1996

How good is the CLAMS vegetation map?

• Assessed accuracy using a variety of methods

• Excellent representation of regional patterns and variability, landscape proportions

• Reasonable representation of fine-scale pattern, inexact for specific sites, similar to other satellite-based maps

• Rare species and habitats not well represented

• For more information:

– Posters

– Ohmann, J.L.; Gregory, M.J. 2002. Predictive mapping of forest composition and structure with direct gradient analysis and nearest neighbor imputation in coastal Oregon, USA. Canadian Journal of Forest Research 32:725-741.

Uses of Vegetation Map in CLAMS

• Initial conditions (1996) for landscape simulations

• Response models for wildlife, aquatic, timber

• ‘Big picture’ vegetation conditions

• Current vegetation biodiversity

Current policy

Alt A

Alt B

Alt C

Natural Processes

Landowner Behavior

t =1BiophysicalResponse t =n

Landscape/ Watershed Condition

t =1 t =n

Socio-economicResponse t =1

Coast Range Ecosystem

t =n

CLAMS conceptual model

Forest Types and Management Objectives• About 1/3 of each forest type managed for ecological goals

EXCEPT...

• Foothill oak woodlands: 94% on private lands, few reserves, threatened by nonforest development.

0%

20%

40%

60%

80%

100%

Sitka spruce Westernhemlock

P. silverfir/noble fir

Dryw.hemlock/

mixedevergreen

Foothill oakwoodlands

Timber primary

Timber and othergoals

Ecological primary,timber secondary

Ecological only

No-harvestreserves

Timbergoals

predominate

Ecologicalgoals

predominate

Natural legaciesafter wildfire

Lack oflegacies under intensive management

Forest management w/ legacies

Legacy Trees

Key Findings: Vegetation Biodiversity in Coastal Oregon

• In semi-natural forested landscapes, all ownerships contribute to biodiversity.

• Some biodiversity elements (tree species, forest types) are relatively insensitive to forest management practices: conservation must consider regional environmental gradients.

• Forest types represented in reserves EXCEPT foothill oak woodlands.

• Older forests: small part of current landscape and below HRV, but being addressed by current policies. Diverse young forests: also rare but receiving less attention. Legacy tree habitat: uncertain future.

What’s so novel about the CLAMS vegetation map?(i.e., advantages for ecological analysis,

simulation modeling, integrated assessment)

• Spatially complete, regional in scope, AND rich in detail (tree species and structures)

• Each pixel contains a tree list, from which many continuous vegetation variables can be derived. User-defined classification systems can be applied to meet a variety of objectives.

• At regional level, full range of variability is represented. At site level, covariance of species and structures is maintained.

• Use of mapped environmental data results in models that better capture ecological relationships.